RETRACTED ARTICLEL: An automated neural network based technique for identifying fundus hemorrhage (NNTFH)

Abstract

Retina, the thin membranous tissue layer occupying the back of human eyes, provides vision to humans. As the age of a person increases the eyes may encounter a secondary growth, creating impairments in vision. Human eyes are prone to several diseases like retinal detachment or tear, glaucoma, macular degeneration or hole, diabetic retinopathy etc., where identifying retinal diseases at an early stage is necessary. The increasing number of eye affected patients and effective diagnosis imposes a challenge on the clinical routines of treatment and monitoring after diagnosis. It is possible to diagnose eye diseases from retinal images with the help of machine learning techniques. This paper proposes a novel technique called NNTFH which is an automated neural network based technique for identifying hemorrhage of the eyes from Eye images. The initial phase of NNTFH, selects the pixel count and density from medical eye images and then classifies impaired eyes where it uses a neural network model. The retinal images are classified as normal or with exudates or eyes with Hemorrhage.